Design And Implementation Of Intelligent Voice Interactive Chart Generation System

Chart visualization system is the most effective and concise way of data analysis.However,the visualization system based on button interaction limits the flexibility and intelligence of the system.With the continuous progress and development of in-depth learning,natural language processing technology and speech processing technology have achieved an all-round development.The research on natural language-generated SQL,nl2 sql,has been developing rapidly.Nl2 sql technology and voice technology should be applied to the visualization system to effectively improve the flexibility of the system.Therefore,this paper applies nl2 sql technology and voice processing technology to the chart visualization system,and designs and implements an intelligent voice interactive chart generation system.There are many challenges in the implementation of the system.In the face of different challenges,first,the system implements the speaker recognition algorithm by using the dynamic time regulation algorithm DTW and the unbounded staggered state regression neural network UIS-RNN.Second,the nl2 sql model is improved by using BERT and prior knowledge,and the model are optimized by combining application scenarios.Then,the system uses support vector machine SVM,TF-IDF design and implementation of the chart type matching algorithm;In addition,the Spring Boot technology is used to complete the development of the system.The system is divided into five modules,the data management module is used to manage and maintain the data;The voice control module is used for user authentication and real-time speech recognition;According to the commands of speech recognition,the data retrieval module automatically generates SQL according to the commands and completes data retrieval;Chart visualization module is used to match visualized chart types according to voice commands and to visualize data;User management module is used to maintain and manage user information.This system is based on the visual analysis of academic data.Intelligent voice retrieval improves the flexibility of data retrieval;A rich variety of graphs can be adapted to different analysis scenarios.The system as a whole greatly improves the efficiency of data analysis and helps users to obtain the value of data.